Skip to content

qauntize cli: Add gptq as supported implementation #2115

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
jambayk merged 1 commit into from
Aug 20, 2025
Merged

qauntize cli: Add gptq as supported implementation #2115

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 2 additions & 1 deletion olive/cli/quantize.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -133,7 +133,7 @@ def _get_pass_list(self, precision, algo, impl, is_hf_model):

if not pass_list:
raise ValueError(
f"Quantiation for precision {precision}, algorithm {algo} and implementation {impl} "
f"Quantization for precision {precision}, algorithm {algo} and implementation {impl} "
f"with QDQ {self.args.use_qdq_encoding} is not supported"
)

Expand Down Expand Up @@ -237,6 +237,7 @@ def run(self):
# TODO(jambayk): consider exposing the activation bits through the act_precision argument
{"impl_name": ImplName.SPINQUANT, "pass_type": "SpinQuant"},
{"impl_name": ImplName.AWQ, "pass_type": "AutoAWQQuantizer"},
{"impl_name": ImplName.OLIVE, "pass_type": "Gptq"},
{"impl_name": ImplName.AUTOGPTQ, "pass_type": "GptqQuantizer"},
]

Expand Down
8 changes: 6 additions & 2 deletions olive/common/ort_inference.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -37,10 +37,14 @@ def ort_supports_ep_devices() -> bool:

def maybe_register_ep_libraries(ep_paths: dict[str, str]):
"""Register execution provider libraries if onnxruntime supports it."""
if not ort_supports_ep_devices():
try:
import onnxruntime as ort
except ImportError:
logger.debug("Skipping EP registration since onnxruntime is not installed")
return

import onnxruntime as ort
if not ort_supports_ep_devices():
return

# providers that ort was built with such as CUDA, QNN, VitisAI but need registration
for provider in set(ort.get_available_providers()):
Expand Down